Re: Version identifier in URL
Daniel W. Connolly (connolly@beach.w3.org)
Wed, 29 May 1996 18:50:01 -0400
In message <199605290508.WAA05273@spice.p3.com>, Christopher Seiwald writes:
>I'm not sure I understand Dan Connolly's objection to standardizing
>a version extension to URLs, other than that he is objecting to it.
URLs are supposed to be opaque strings that identify resources.
Information providers may encode semantic information in them, but
clients must treat them as opaque, except by private agreement.
Peeking into the syntax of a URL is an abstraction violation.
Telling folks "watch out! Don't spell your URLs this way, cuz
that syntax is reserved for version info" is a no-no.
What if somebody is already using the syntax ,xxx for
some other purpose? According to the URL standard, they're allowed
to do that.
Standardizing ,xxx for version info would be like standardizing .ps
for postscript resources. That's just not how it works. (the mapping
of filename extension to media type is a per-server convention. Folks
who exploit it on the client side do so on a heuristic basis, at their
own peril)
>It does underscore the fact that attaching version information to URLs
>is going to be a battle, but a battle that I believe is well worth
>fighting.
I assure you that Tim Berners-Lee and myself will do everything we
can to prevent you from winning.
Please see the original writing on the topic of versions in WWW:
--------------
HyperText Design Issues: Versioning
http://www.w3.org/pub/WWW/DesignIssues/Versioning.html
A WWW server may provide versioning, by allowing links between a
document version and its previou and succesive versions. This would be
a good use of link typing.
Keeping track of versions allows one solution to the annotation problem.
--------------
and more recently:
Web Architecture: Generic Resources
http://www.w3.org/pub/WWW/DesignIssues/Generic
This is one of the few things that TimBL felt sufficiently strongly
about to take time out of his busy schedule and write about.
If you do not understand my objection, I suggest you study the
design and architecture of WWW and URLs until you do.
>We need to stick to our guns: make a simple version extention to URLs
>and make them standard.
I am confident that no such standard will be issued with the backing
of W3C while Tim Berners-Lee is director.
Daniel W. Connolly "We believe in the interconnectedness of all things"
Research Scientist, MIT/W3C PGP: EDF8 A8E4 F3BB 0F3C FD1B 7BE0 716C FF21
<connolly@w3.org> http://www.w3.org/pub/WWW/People/Connolly/